Perceptions, Characteristics and Behaviors of Cigarette and Electronic Cigarette Use among Pregnant Smokers

Abstract

Introduction

Smoking during pregnancy is a primary risk factor for adverse perinatal outcomes. While electronic cigarette (e-cigarette) use has increased, reasons for and behaviors of use are not fully understood. The purpose of this study, comprised exclusively of pregnant current smokers, was to describe perceptions of health risks associated with e-cigarette use among pregnant women, describe the use patterns of pregnant dual users (defined as those who simultaneously use conventional/combustible cigarettes and e-cigarettes), and examine smoking-related behaviors between conventional-only (defined as those smoking combustible cigarettes but not e-cigarettes) and dual users.

Methods

Cross-sectional data from a larger study of pregnant conventional-only and dual users was analyzed. A brief survey measured perceptions of prenatal e-cigarette use and smoking behaviors. Analysis included descriptive statistics, bivariate analysis, and logistic and linear regression analysis using SAS v9.4 (alpha .05).

Results

Among 176 pregnant smokers (38% dual users), more than half of participants believed e-cigarettes were harmful to women (56%) and posed harm to the fetus (53%). Among dual users, 41% used their e-cigarette daily, on average eight times/day. Eleven percent of dual users smoked a pack of cigarettes/day, compared to 5% of conventional-only smokers, and dual users scored significantly higher (p=.026) on the Penn State Cigarette Dependence Index. The most common e-cigarette liquid flavor was fruit (64%), and the most frequently reported e-cigarette nicotine concentration was 1–6mg.

Conclusion

Characterizing perceptions and behaviors of e-cigarette use during pregnancy is foundational for future research to explore the association between product use and maternal and infant outcomes.

Introduction

Cigarette smoking is associated with poor health outcomes including cancer and heart disease, and remains the leading cause of preventable deaths in the U.S. (Centers for Disease Control (CDC), 2019). Nearly 38 million (15.5%) U.S. adults, and more specifically, 13.5% of women currently smoke cigarettes (CDC, 2019). Despite decades of scientific evidence linking smoking during pregnancy to adverse perinatal outcomes including miscarriage, low birth weight, preterm birth, and Sudden Infant Death Syndrome (SIDS) (U.S. Department of Health and Human Services (USDHHS), 2014), 7.2% of U.S. women who gave birth in 2016 reported smoking during pregnancy, and states such as West Virginia and Kentucky have prenatal smoking rates that greatly exceed the national rate (25.1% and 18.4% respectively) (Drake, P., Driscoll, A., and Mathews, T., 2018).

Further complicating the high rates of tobacco use in pregnancy is the 2007 emergence of electronic nicotine delivery systems (ENDS), also known as electronic cigarettes (e-cigarettes), to the U.S. market. E-cigarettes are battery-operated smoking devices with various names including JUUL, mods, vape pens, etc., and are available in many brands, sizes, and voltages. E-cigarettes operate by using a battery to heat liquid (common ingredients include nicotine, glycerin, propylene glycol, and flavor agents) in the e-cigarette cartridge or tank to produce an aerosol that is inhaled by the user (USDHHS, 2016). E-cigarettes contain nicotine, a highly addictive constituent of tobacco products that is known to increase the risk for adverse fetal outcomes such as preterm birth and suspected to contribute to small for gestational age infants (USDHHS, 2014; England L., Bunnell R., Pechacek T., Tong V., McAfee T., 2016; Dempsey, D., & Benowitz, N., 2001). Most available studies assessing e-cigarette use do not measure e-cigarette nicotine content (Weaver, S., Kim, H., Glasser, A., Sutfin, E., Barrington-Trimis, J., Payne, T. et al., 2018); however, a few studies have indicated the level of nicotine delivered from e-cigarettes is similar to or higher than levels from conventional cigarettes (St Helen, G., Havel, C., Dempsey, DA et al., 2016; Etter, J., 2014). Plasma nicotine concentrations among e-cigarette users have been demonstrated to be similar to those among conventional cigarette users (USDHHS, 2016). This poses a significant public health concern for maternal and child health, as any level of nicotine is known to be harmful to a developing fetus.

E-cigarette use continues to evolve as new devices emerge and products gain in popularity. A recent analysis of data from the Population Assessment of Health and Tobacco (PATH, collected in 2013–2014) survey concluded 5.9% of women of reproductive age (15–44 years) were current e-cigarette users (Lopez et al., 2018). Kurti and colleagues (2017) found 4.9% of pregnant respondents in the PATH survey were current e-cigarette users, with most of them concurrently using conventional cigarettes (a phenomenon known as dual use). E-cigarette use in pregnancy has been attributed to perceptions of e-cigarettes as a safer alternative to conventional cigarettes and as an aid for smoking cessation (Fallin, Miller, Assef & Ashford, 2016; Kahr et al., 2015; Mark, Farquhar, Chisolm, Coleman-Cowger, and Terplan, 2015; McCubbin, Fallin-Bennett, Barnett & Ashford, 2017); however, e-cigarettes have not been shown to be an effective method for smoking cessation during pregnancy (Siu, Al., 2015). Research examining e-cigarette use among pregnant women is scant. In fact, a search of the ClinicalTrials.gov database (an online registry of active and closed clinical trials) in November 2019, using the search terms “electronic cigarette” AND “pregnant”, yielded only two U.S. studies (in addition to this study) currently recruiting pregnant participants.

Dual use is a common behavior for e-cigarette users and may compound adverse health and fetal outcomes associated with conventional cigarette use (Marufu, T., Ahankari, A., Coleman, T., Lewis, S., 2015; CDC, 2018). During pregnancy, women may attempt to switch from conventional cigarettes to e-cigarettes to lessen the perceived harm to the fetus, or in an attempt to assist with cessation of conventional cigarettes (Peterson, L. & Hecht, S., 2017; Adkison, S., O’Connor, R., Bansal-Travers, M., Hyland, A., Borland, R., Yong, H. et al., 2013). Speculation exists that dual use occurs during the process of switching because, when converting to e-cigarettes, the smoker may struggle to achieve the same ‘hit’ and therefore supplement with a conventional cigarette.

The purpose of this study was to describe perceptions of health risks and acceptability associated with e-cigarette use in pregnancy, describe the use patterns of pregnant dual users, and compare smoking-related behaviors between pregnant conventional-only and dual users. To our knowledge, this study is the first to examine these findings in a sample comprised exclusively of pregnant conventional and dual smokers with current tobacco use.

Methods

This study was part of a larger, on-going, multicenter study evaluating the impact of electronic cigarette use on perinatal immune response among women with singleton gestations receiving prenatal care in one of three obstetric clinics in central Kentucky. Patients receiving care in the three clinics were predominantly white, and included a mix of privately and publicly insured patients, as well as some without insurance. At enrollment, participants were 18–44 years of age, pregnant and in first or second trimester, able to read/write in English, and had used cigarettes and/or e-cigarettes within the past 30 days. Cross-sectional enrollment data from part of that larger study were evaluated here. The study was approved by the University of Kentucky Institutional Review Board (IRB). A research nurse trained in human subjects protections explained the study to eligible patients and obtained voluntary informed consent. In addition to self-report of tobacco use, which has been linked to misclassification during pregnancy (Ashford, K., Wiggins, A., Rayens, E., Assef, S., Fallin, A. & Rayens, MK., 2017), tobacco use was biochemically validated at enrollment using Nymox NicAlert™ urine strips with preset cotinine limits ranging from 0 (nonsmoker) to 6 (cotinine concentration ≥2,000 ng/mL) (Gaalema, Higgins, Bradstreet, Heil & Bernstein, 2011); however, this measure was not a criterion for study exclusion. Participants completed a 20-minute self-administered survey (written at a sixth grade level) on a clinic iPad. Data were recorded in REDCap, a secure web-based data management system managed through the University of Kentucky Center for Clinical and Translational Science (CCTS). Participants received a $25 Walmart gift card at enrollment.

Measures

Demographic information collected via survey included age, race, ethnicity, education, income, and employment status. Participants were asked, ‘Which of the following best describes your race?’ with response options including ‘American Indian/Alaskan Native,’ ‘Asian,’ ‘Native Hawaiian or Other Pacific Islander,’ ‘Black or African American,’ ‘White,’ and ‘More than 1 race.’ Responses were combined and a dichotomous variable (‘White’ or ‘Non-white’) was used in subsequent analyses. For education, women were asked ‘What is the highest grade or year of school you have completed?’ with response options including ‘Less than high school graduate,’ ‘High school graduate or GED,’ ‘Some college or vocational/trade school,’ and ‘College graduate or beyond.’ Employment for wages was categorized as yes or no (yes =‘part-time’ or ‘full-time’; no = ‘unemployed,’ ‘student’ or ‘homemaker’). Participants were asked to choose their annual household income from response options including ‘<$20,000,’ ‘$20,000-$49,999,’ or ‘$50,000 or more’.

Cigarette user groups were determined based on participant answers to two questions in the survey: 1) ‘Have you smoked cigarettes at any time during the last 30 days?’ and 2) ‘Have you used e-cigarettes within the past 30 days?’ Instructions within the survey text reminded participants ‘E-cigarettes are also called vapes, mods, hookahs, vape pens, and JUULs and can be disposable or re-usable. When we say e-cigarette we are referring to all of these types of products.’ Respondents who answered ‘yes’ to cigarette use in the past 30 days and ‘no’ to e-cigarette use in the same time period were categorized as conventional-only users. Those who answered ‘yes’ to both cigarette and e-cigarette use in the past 30 days were classified as dual users.

Perceptions associated with use of e-cigarettes were assessed by participant responses to 14 questions measuring general health risks, acceptability, and perceived risks to specific health scenarios (e.g. maternal health, cancer, lung disease, fetal health) (see Table 1). Respondents’ beliefs were analyzed overall, and then between conventional-only and dual users. Acceptability of e-cigarette use was measured with the question, ‘In your opinion, how acceptable is it for [group] to use e-cigarette?’ with ‘group’ defined as women in general, pregnant women, and postpartum women (had a baby in the past 10 weeks). Response options included ‘not at all acceptable,’ ‘somewhat acceptable,’ ‘acceptable,’ or ‘very acceptable’ (adapted from Nichols, T. R., Birnbaum, A. S., Birnel, S., & Botvin, G. J., 2006).

Table 1.

Survey questions and associated response options to measure perceptions, acceptability and health risks of e-cigarette use

Survey questions	Answer choices
Perceived risk of e-cigarettes (adapted from Bhanji, S., et al., 2011)
In general, do you feel the e-cigarette is a:	○ Serious health hazard
	○ Moderate health hazard
	○ Minor health hazard
	○ Not a health hazard at all
	○ Don’t know
Perceived risk of e-cigarette to women’s health (adapted from Bhanji, S. et al., 2011)
Please check no or yes to each item.
a) Is e-cigarette use harmful to user?	○ No
b) Does e-cigarette use cause cancer?	○ Yes
c) Does e-cigarette use cause heart disease?	○ Don’t know
d) Does e-cigarette use cause lung disease?
e) Does e-cigarette use have harmful effects on fetus?
f) Does passive e-cigarette exposure have harmful effects on children (passive means to inhale smoke/vapor from someone else’s e-cigarette use)?
g) Does occasional e-cigarette use have harmful effects on the user?
h) Is e-cigarette use associated with infertility?
i) Does e-cigarette use relieve stress?
j) Does e-cigarette use help in losing weight?
Acceptability of e-cigarette use (adapted from Nichols, T.R. et al., 2006)
In your opinion, how acceptable is it:	○ Not at all acceptable
a) For women to use e-cigarettes	○ Somewhat acceptable
b) For pregnant women to use e-cigarettes	○ Acceptable
c) For post-partum women (had a baby in the past 10 weeks) to use e-cigarettes	○ Very acceptable

Perceptions of the health risks of using e-cigarettes were measured by responses to 11 questions and analyzed overall and between conventional-only and dual users. General health risk was assessed by asking ‘In general, do you feel the e-cigarette is a ‘Serious health hazard,’ ‘Moderate health hazard,’ ‘Minor health hazard,’ ‘Not a health hazard at all,’ or ‘Don’t know’ (adapted from Bhanji, S., Andrades, M., Taj, F., & Khuwaja, A. K., 2011). Respondents were then asked ten ‘Yes/No’ questions about specific risks of e-cigarette use to women’s health, including perceived risk as ‘harmful to the user,’ ‘harmful to the fetus,’ and ‘causes cancer’ (adapted from Bhanji, S. et al., 2011) (see Table 1).

Behaviors of e-cigarette use were measured using survey items assessing frequency and duration of use, nicotine concentration in e-cigarette liquid, flavors of liquid, and reason for e-cigarette use. Frequency was gauged by responses to two open-ended questions: ‘How many days in the past 30 days have you used an electronic cigarette?’ and ‘How many times PER DAY do you usually use your electronic cigarette? Assume one “TIME” usually consists of 15 puffs or lasts around 10 minutes’ (Foulds et al., 2014). The number of days a participant used e-cigarettes was dichotomized into daily and non-daily for analysis (daily=30 days; non-daily ≤29).

Participants were also asked ‘When was the last time you used an e-cigarette?’ with response options in four categories ranging from ‘within the past day/24 hours’ to ‘more than 3 days ago.’ Nicotine concentration of e-cigarette liquid was measured by asking dual users ‘What concentration of nicotine was in your most recent e-cigarette?’, with six response categories ranging from ‘0mg’ to ‘25mg or more.’ Current e-cigarette liquid flavors preferred by dual users were determined by summarizing responses to ‘What flavors do you currently use [in your e-cigarette]?’ with choices including ‘mint/menthol,’ ‘clove or spice,’ ‘fruit,’ ‘chocolate,’ ‘alcoholic drink (such as wine, cognac, margarita or other cocktails),’ ‘candy or other sweets,’ ‘some other flavor (describe),’ or ‘don’t know.’ To determine reason for e-cigarette use, the participants were asked ‘Why did you decide to use e-cigarettes? Check all that apply’ and were provided thirteen answer options including ‘I became pregnant,’ ‘less harmful to myself,’ and ‘wanted to quit smoking’ (Figure 1).

Figure 1.

Reasons for deciding to use e-cigarette among pregnant dual users (n = 66)

Similar to e-cigarette behaviors, conventional cigarette use behaviors were determined from responses to ‘Have you smoked at least 100 cigarettes in your lifetime?’ (yes/no), and two open-ended questions, ‘How many days in the past 30 days have you smoked conventional cigarettes?’ and ‘Thinking of the last 30 days, how many cigarettes did you smoke per day?’ The number of days a participant used cigarettes was dichotomized into daily and non-daily for analysis (daily=30 days; non-daily ≤29).

The 10-item Penn State Cigarette Dependence Index (PSCDI) (index scores range 0–20) was used to measure current dependence (Foulds et al., 2014). PSCDI questions included ‘How many cigarettes per day do you usually smoke?’, ‘Do you sometimes awaken at night to have a cigarette?’, and ‘Do you ever have strong cravings to smoke?’ Answers were summed, with the total score corresponding to a dependence category: Not dependent (0–3), Low dependence (4–8), Medium dependence (9–12), and High dependence (13+).

For quitting patterns, women were asked, ‘How many times have you attempted to quit smoking cigarettes?’ Because the distribution of quitting attempts was right skewed, quitting was dichotomized based on a median split. The median was three quit attempts, so the categories were three or fewer quit attempts versus more than three attempts for analysis purposes.

Data Analysis

Descriptive statistics, including frequency distributions, medians and interquartile ranges, or means and standard deviations, were used to summarize study variables. Bivariate analysis methods, including the two sample t-test, chi-square test of association, and the Mann-Whitney U test were used to determine differences in demographics, perceptions of e-cigarettes, and smoking behaviors between conventional-only and dual users. Logistic regression was used to assess the relationship between tobacco use group and dichotomous outcomes (daily/non-daily cigarette use, more than three quit attempts/three or fewer quit attempts), controlling for age, race/ethnicity, education, income, and employment status. In addition to these control variables, daily cigarette use and cigarettes per day were included as covariates in the quitting attempts logistic regression model. Ordinal logistic regression was used to assess the adjusted relationship between tobacco use group and cigarettes per day (less than 10, 10–20, more than 20), and linear regression evaluated the adjusted relationship between tobacco use group and nicotine dependence. Data analysis was conducted using SAS, version 9.4 (Cary, NC), with an alpha level of .05 throughout.

Results

Sample Profile

From January 2016 to May 2019, 197 pregnant smokers meeting eligibility criteria were enrolled. Study participation was voluntary and the electronic survey did not require a response to every question, thus response rates varied by question. Data on tobacco use and perceptions of e-cigarettes were available for 176 pregnant tobacco users (110 conventional-only; 66 dual). At the time of analysis, only 12 exclusive e-cigarette users were enrolled; they were excluded from this analysis due to small sample size. The average age of study participants was 28.3 years (SD=5.6; Table 2). The majority identified their race/ethnicity as White, non-Hispanic (83%) and had at least a high school education (86%). Approximately half were unemployed (52%) and almost two-thirds (62%) reported a household income less than $20,000. There was no statistically significant difference in age, race/ethnicity, education, income, or employment between dual and conventional-only users.

Table 2.

Demographic characteristics by tobacco use group (conventional-only and dual)

	Total sample (N = 176)	Conventional-only (n =110)	Dual user (n = 66)	p
	Mean (SD) or n (%)	Mean (SD) or n (%)	Mean (SD) or n (%)
Age	28.3 (5.6)	28.2 (5.5)	28.6 (5.8)	.62a
Race/ethnicity				.49b
White, non-Hispanic	145 (82.9%)	92 (84.4%)	53 (80.3%)
Hispanic or other race	30 (17.1%)	17 (15.6%)	13 (19.7%)
Education				.32c
Less than high school	24 (13.7%)	17 (15.6%)	7 (10.6%)
High school	73 (41.7%)	46 (42.2%)	27 (40.9%)
More than high school	78 (44.6%)	46 (42.2%)	32 (48.5%)
Household income				.75c
Less than $20,000	95 (61.7%)	58 (62.4%)	37 (60.7%)
$20,000 – $49,999	39 (25.3%)	24 (25.8%)	15 (24.6%)
$50,000 or more	20 (13.0%)	11 (11.8%)	9 (14.8%)
Employed for wages				.92b
Yes	84 (48.0%)	52 (47.7%)	32 (48.5%)
No	91 (52.0%)	57 (52.3%)	34 (51.15%)

^ap from two-sample t-test

^bp from chi-square test of association

^cp from Mann-Whitney U test

Perceptions of Health Risks and Acceptability of E-cigarette Use among Pregnant Smokers

In the overall sample, 30% of respondents (n = 54) indicated perceived harm of e-cigarette use as ‘no health hazard at all’ or ‘minor health hazard,’ and 20% responded ‘moderate health hazard’ (n = 24) or ‘serious health hazard’ (n = 12); however, 49% of the sample chose the answer ‘don’t know’ when considering the health hazard of e-cigarette use (n = 88; Table 3). Interestingly, among those who chose an option other than ‘don’t know,’ significantly more dual users perceived e-cigarettes as being a health hazard compared to conventional-only users (p <.001).

Table 3.

Perceptions of e-cigarette among pregnant smokers

	Total sample (N = 176) n (%)	Conventional only (n =110) n (%)	Dual user (n = 66) n (%)	p
Perceived risk of e-cigarette
General thoughts regarding e-cigarette:				<.001a
Serious health hazard	12 (6.8%)	8 (7.2%)	4 (6.0%)
Moderate health hazard	24 (13.6%)	18 (16.2%)	6 (9.0%)
Minor health hazard	41 (23.3%)	15 (13.5%)	26 (38.8%)
Not a health hazard at all	13 (7.4%)	6 (5.4%)	7 (10.4%)
Don’t know	86 (48.9%)	64 (57.7%)	24 (35.8%)
Perceived risk of e-cigarette to women’s health	% yes	% yes	% yes	pb
E-cigarette harmful to user	56.2%	63.0%	45.0%	.024
E-cigarette cause cancer	35.6%	40.8%	26.8%	.081
E-cigarette cause heart disease	33.3%	39.4%	23.6%	.050
E-cigarette cause lung disease	48.6%	52.6%	42.6%	.24
E-cigarette have harmful effects on fetus	53.3%	60.4%	41.2%	.027
E-cigarette have harmful effects on children	44.4%	46.5%	42.1%	.60
Occasional e-cigarette use harmful to user	42.3%	46.1%	36.7%	.24
E-cigarette use is associated with infertility	19.9%	23.9%	13.0%	.11
E-cigarette use relieves stress	82.4%	81.0%	82.8%	.77
E-cigarette use helps in losing weight	31.6%	35.7%	25.4%	.18
Acceptability
For women to use e-cigarette				.002c
Not at all acceptable	10 (5.8%)	8 (7.0%)	2 (3.0%)
Somewhat acceptable	27 (15.5)	23 (20.2%)	7 (10.6%)
Acceptable	85 (48.8%)	58 (50.9%)	29 (43.9%)
Very acceptable	52 (29.9%)	25 (21.9%)	28 (42.4%)
Pregnant women to use e-cigarette				.001c
Not at all acceptable	35 (20.5%)	29 (26.1%)	7 (10.6%)
Somewhat acceptable	68 (39.8%)	47 (42.3%)	24 (36.4%)
Acceptable	53 (31.0%)	29 (26.1%)	26 (39.4%)
Very acceptable	15 (8.8%)	6 (5.4%)	9 (13.4%)
Postpartum women to use e-cigarette				.002c
Not at all acceptable	16 (9.2%)	14 (12.4%)	3 (4.6%)
Somewhat acceptable	49 (28.3%)	36 (31.9%)	15 (22.7%)
Acceptable	70 (40.5%)	47 (41.6%)	26 (39.4%)
Very acceptable	38 (22.0%)	16 (14.2%)	22 (33.3%)

^ap from Mann-Whitney U test, omitting those who responded “Don’t know”

^bp from chi-square test of association

^cp from Mann-Whitney U test

In contrast to the minority of participants who considered e-cigarette use a moderate or serious health hazard (as noted above), the majority of participants agreed that e-cigarettes were harmful to women (n = 95, 56%), have harmful effects to the fetus (n = 76, 53%), and cause lung disease in women (n = 74, 49%). The majority of participants also concurred that e-cigarette use relieves stress (n = 134, 82%). Significant differences were revealed, however, when analyzing perception of risk to women’s health by tobacco user group. Contrary to the incongruity between use groups above regarding perceptions of e-cigarette risk on general health, a higher proportion of conventional-only users, compared to dual users, agreed that e-cigarettes are harmful to women (63% vs. 45%, p=.024), pose harmful effects to the fetus (60%, vs. 41%, p=.027), and cause heart disease in women (39% vs. 24%, p=.050).

Examining acceptability of e-cigarette use among all smokers (conventional-only and dual) revealed the majority (n = 137, 79%) thought it was ‘somewhat acceptable’ or ‘very acceptable’ for women to use e-cigarettes (see Table 3). While more than half (n = 108, 63%) of all smokers thought it was ‘somewhat acceptable’ or ‘very acceptable’ for postpartum women to use e-cigarettes, far fewer (n = 68, 39%) selected these categories of e-cigarette use acceptability for pregnant women. As expected, when we compared perceptions of acceptability of e-cigarette use between conventional-only users and dual users, the dual users perceived e-cigarette use as more acceptable for women (p=.002), pregnant women (p=.001), and postpartum women (p=.002), respectively.

Use Behaviors of Pregnant Dual Users

Of the 66 pregnant dual users, the majority reported last using their e-cigarette within the past 24 hours (n = 38, 61%), while the remaining participants reported last using their e-cigarette within the past two days (n = 5, 8%), three days (n =4, 6%), or more than 3 days ago (n = 15, 24%; see Table 4). The most frequently reported nicotine concentration range used in their most recent e-cigarette was 1mg – 6mg (n = 33, 56%). Less than half of the women reported using their e-cigarette daily (n = 26, 41%), while others were equally divided between less than ten days a month (n = 18, 29%) and 10–15 days per month (n = 18, 29%). The median number of times per day of e-cigarette use was 8 times (IQR = 3 – 15). Among those who used flavors, the most commonly selected flavor was fruit (n = 42, 64%), followed by mint/menthol (n = 23, 35%), and candy flavors (n = 18, 27%). The most frequently selected reasons for deciding to use e-cigarettes were ‘wanted to quit smoking’ (n = 55, 83%), ‘less expensive’ (n = 32, 49%), ‘less harmful to self’ (n = 30, 46%), and ‘like the taste/flavors’ (n = 28, 42%) (Figure 1).

Table 4.

Patterns of e-cigarette use among pregnant dual users (n = 66)

Characteristic	n (%)
Last time used e-cigarette
Within the past day	38 (61.3%)
Within the past 2 days	5 (8.1%)
Within the past 3 days	4 (6.4%)
More than 3 days ago	15 (24.2%)
Concentration of nicotine in most recent e-cigarette
0 mg	5 (8.2%)
1 mg – 6 mg	33 (54.1%)
7 mg – 12 mg	10 (16.4%)
13 mg – 18 mg	3 (4.9%)
19 mg – 25 mg	5 (8.2%)
25 mg or more	3 (4.9%)
Decline to answer	2 (3.3%)
Number of days used an e-cigarette in past 30 days
Less than 10 days	18 (28.6%)
10–15 days	18 (28.6%)
16–29 days	3 (4.8%)
Daily	26 (41.3%)
Number of times per day of e-cigarette use, median (interquartile range)	8 (3 – 15)
E-cigarette flavors currently using (select all that apply; % checked)
Mint/menthol (n = 23)	34.9%
Clove/spice (n = 2)	3.0%
Fruit (n = 42)	63.6%
Chocolate (n = 2)	3.0%
Alcoholic drink flavor (n = 1)	1.5%
Candy or other sweets (n = 18)	27.3%

Note: numbers vary due to sporadically missing data

Comparison of Smoking Behaviors between Pregnant Conventional Cigarette and Dual Users

In order to compare smoking behaviors between conventional cigarette and dual users, analysis included number of cigarettes smoked per day (CPD) and cigarette dependence scores (PSCDI). Compared to dual users, a higher proportion of conventional-only users reported smoking cigarettes daily (90% vs. 70%, adjusted and unadjusted p’s <.001; see Table 5). More dual users (n =7, 11%) than conventional-only users (n = 6, 5%) reported smoking more than a pack per day (>20 CPD), however the difference between the two groups was not statistically significant. Dual users scored significantly higher on the Penn State Cigarette Dependence Index (M =11.8, SD=4.7 vs. M =10.6, SD=3.6, p=.026), but this difference was no longer significant after adjusting for age, race/ethnicity, education, income, and employment status. There was no significant difference in quitting patterns (having previously tried to quit more than three times versus three or fewer times) between dual and conventional only users.

Table 5.

Adjusted and unadjusted associations among tobacco use group and smoking related behaviors

	Conventional only users (n = 110) n (%)	Dual users (n = 66) n (%)	Unadjusted p	Parameter estimates	adjusted p
Frequency of use (conventional)			<.001	OR = 4.68; 95% CI = (1.89 – 11.58)	<.001a
Daily cigarette smoker	107 (89.9%)	46 (69.7%)
Less than daily	12 (10.1%)	20 (30.3%)
Cigarettes per day			.53	OR = 1.63; 95% CI = (0.79 – 3.34)	.32b
<10	69 (58.0%)	36 (55.4%)
10–20	44 (37.0%)	22 (33.8%)
>20	6 (5.0%)	7 (10.8%)
Penn State nicotine dependence index	10.6 (3.6)	11.8 (4.7)	.026	b = −1.11; SE = 0.64	.084c
Quit attempts	57 (60.0%)	26 (45.6%)
Three or fewer	38 (40.0%)	31 (54.4%)	.085	OR = 0.65; 95% CI = (0.30 – 1.38)	.26a
More than three

Note: Models were adjusted for age, race/ethnicity, education, income, and employment status; the quit attempts model is also adjusted for daily and cigarettes per day.

^ap from logistic regression

^bp from ordinal logistic regression

^cp from linear regression

Discussion

To our knowledge, this study is the first to describe perceptions of health risks and acceptability of use associated with e-cigarette use in pregnancy, describe the use patterns of pregnant dual users, and compare smoking-related behaviors between pregnant conventional-only and dual users in a sample comprised exclusively of pregnant current smokers. While a limited number of studies have examined perceptions and behaviors of e-cigarette use among women, many of these studies included perceptions of former-smokers, never-smokers, never-e-cigarette users, and/or non-pregnant female participants (Bhandari et al, 2018; Bowker et al, 2018; England, L., Tong, V., Koblitz, A., Kish-Doto J., Lynch, M., Southwell, B., 2016; Ashford, K., Wiggins, A., Butler, K., Ickes, M., Rayens, M., Hahn, E., 2016; Wagner, N., Camerota, M., Propper, C., 2017). For example, while Bhandari and colleagues (2018) surveyed 382 pregnant women to determine knowledge, use, and perceptions of tobacco and e-cigarettes, 69% of the participants had never used e-cigarettes, 37.3% had never used e-cigarettes nor tobacco, and only 11.9% reported current e-cigarette use. Similarly, Mark and colleagues (2015) conducted research with 316 pregnant women to evaluate knowledge and attitudes of e-cigarette use among pregnant women, yet only 0.6% of study participants were current e-cigarette users. In addition, Kurti et al. (2017) summarized prevalence and correlates of conventional cigarette, e-cigarette, and other nicotine delivery product use among 388 pregnant women in the PATH survey; however, only a very small proportion were conventional tobacco or e-cigarette users (13.8% and 4.9%, respectively), and less than one-third of the conventional tobacco users reported dual use.

Perceptions of health risks and acceptability of e-cigarette use among pregnant smokers are important factors to understand as researchers explore how the emergence of e-cigarette devices impacts pregnant women. Numerous studies have determined the majority of pregnant women believe e-cigarettes are generally safer than conventional cigarettes (Bhandari et al., 2018; Bowker et al, 2018; Kahr et al., 2015; McCubbin, McCubbin, A., Fallin, A., Barnett, J., Ashford, K., 2017; Wagner, Camerota, & Propper, 2017). Nearly half of the overall sample in the current study indicated they did not know if e-cigarettes were a general health hazard, yet half agreed that e-cigarettes are harmful to women, have harmful effects to the fetus, and cause lung disease. Though significantly more dual users than conventional-only users in the current study believed that e-cigarettes were a health hazard, the conventional-only users indicated that e-cigarettes were harmful to women, posed harmful effects to the fetus, and caused heart disease.

In the current study, the majority of participants perceived e-cigarette use among women as acceptable; however, far fewer reported this behavior as acceptable during pregnancy. The Kahr et al. (2015) study also addressed perceptions of acceptability of e-cigarettes, indicating participants identified smoking in pregnancy as not acceptable yet were less certain about acceptability of e-cigarette use during pregnancy. Additionally, there was a significant difference between conventional-only and dual user groups’ perceived acceptability, with dual users indicating a higher acceptability of e-cigarette use for women, even during pregnancy.

Scientific methods for quantifying e-cigarette use in pregnancy are scant, but a few studies have categorized daily versus non-daily use, defined the number of times per day a person uses their e-cigarette, and examined the nicotine concentration in e-cigarette liquid. As previously summarized, more than half (61%) of the current study reported using their e-cigarette in the past 24 hours, 41% reported daily e-cigarette use with a median use of the device 8 times/day, and the majority (56%) reported 1mg-6mg of nicotine concentration in their most recent e-cigarette. This study is among the first to describe summative use patterns in a sample of pregnant current dual users.

E-cigarettes deliver varying amounts of nicotine (Dawkins and Corcoran, 2014), a known addictive substance, underscoring the importance to measure dependence when examining use behaviors among e-cigarette users. The study by Mark and colleagues (2015) found that 43% of pregnant women who completed a survey about e-cigarette perceptions were unaware that e-cigarettes may contain nicotine. Similarly, analysis of the 2015 Pregnancy Risk Assessment Monitoring System (PRAMS) for Oklahoma and Texas revealed 26.4% of respondents were unsure of the nicotine content of their e-cigarette product (Kapaya et al., 2019).

Dependency

Tobacco dependence is considered a chronic condition (Fiore, M., Jaen, C., Baker, T., Bailey, W., Benowitz, N., Curry, S., et al., 2008) and contributes to increased and prolonged cigarette smoking, as well as difficulty quitting (USDHHS, 2014). Untreated tobacco dependence also increases the risk of developing numerous adverse health problems, including cancer, cardiovascular and respiratory disease, prenatal and newborn complications, weakened immune system, oral disease, and more (CDC, 2018). Understanding dependence is an important element when examining e-cigarette use characteristics, as one-third of the current study’s participants were using both e-cigarettes and conventional cigarettes, and 70% reported daily use of conventional cigarettes. Moreover, 11% of dual users reported consuming more than a pack of cigarettes per day, compared to only 5% of conventional-only users. This finding is significant as it demonstrates that more e-cigarette users in this sample smoked a pack of conventional cigarettes per day compared to participants who were not using e-cigarettes. Further, dual users in the current study scored significantly higher on the Penn State Cigarette Dependence Index compared to conventional-only users (p=.026). Interestingly, 83% of the current study’s dual users indicated they use e-cigarettes to help quit smoking cigarettes, yet adoption of e-cigarettes had not resulted in quitting smoking for these participants at the time of their enrollment in the study.

In 2018, the Public Health Consequences of E-Cigarettes report was released and provided evidence-based conclusions related to outcomes of e-cigarettes. The report identified research gaps related to e-cigarette use in pregnancy, specifically stating, “more research is needed on clinical and epidemiologic studies of e-cigarette use during pregnancy, evaluating the association of patterns of use (including sole and dual e-cigarette use) with maternal and infant outcomes.” (National Academies of Sciences, Engineering, and Medicine. 2018, page 402). Although studies have shown an association of daily and/or nondaily use of conventional cigarettes with increased risk for adverse birth outcomes (Tong, V. T., England, L. J., Rockhill, K. M., & D’Angelo, D. V. ,2017; Mei-Dan, E., Walfisch, A., Weisz, B., Hallak, M., Brown, R., & Shrim, A., 2015), current literature lacks similar analysis for pregnant e-cigarette users and birth outcomes. The current study provides a summary of characteristics for e-cigarette use in pregnancy, which can inform future examinations of potential associations of e-cigarette use with perinatal outcomes.

There are limitations to this study. Although conventional cigarette use was biochemically validated by urine cotinine at study enrollment, survey data related to the number of cigarettes smoked per day and frequency of conventional cigarette and e-cigarette use were collected via self-report, which may lead to misreporting of use or recall bias. Further, the survey question assessing the amount of nicotine concentration used in e-cigarettes did not include a “don’t know” answer option, which may lead to misreporting. The number of pregnant women using e-cigarettes exclusively (no other tobacco product use) at the time of analysis of this population was too small to include (n=12), but is an important area for future research. Finally, this is a small study sample (176 pregnant tobacco users: 110 conventional-only, 66 dual users) and findings may not be generalizable to other prenatal clinics, especially clinics serving areas with a low prevalence of smoking in pregnancy.

Implications for Practice and/or Policy

The findings from this study have public health and clinical implications for prenatal e-cigarette use. All pregnant women should be screened for cigarette use, and providers should specifically ask about e-cigarettes. Clinical care should include clear messaging emphasizing: 1) most e-cigarettes contain nicotine, and there is no safe level of nicotine consumption during pregnancy; 2) increased smoking intensity (which may be compounded by dual use) may lead to increased risk for compounding adverse health and fetal outcomes; and 3) e-cigarette use has not been shown to be an effective method for smoking cessation during pregnancy.

Authors’ Biographies

Andrea McCubbin, MPH¹ is the Director of Perinatal Research Operations for the Perinatal Research and Wellness Center at the University of Kentucky College of Nursing. Her research interests include the impact of tobacco use on perinatal health, and interventions for cessation.

Amanda Wiggins, PhD¹ is a lecturer and statistician for the University of Kentucky College of Nursing. She is the Director of Data Management and Outcomes for UK’s BREATHE team. Her scholarly work primarily focusses on tobacco, adolescent health and perinatal research.

Janine Barnett, MSN¹ is the Nurse Research Coordinator for the Perinatal Research and Wellness Center at the University of Kentucky College of Nursing. She has also worked as a labor and delivery nurse, childbirth educator, and obstetric clinic instructor.

Kristin Ashford, PhD, WHNP-BC, FAAN¹ is the Director of the Perinatal Research and Wellness Center, a Professor, Associate Dean of Undergraduate Faculty Affairs, and Women’s Health Nurse Practitioner at the University of Kentucky College of Nursing.